"""HTML sub-column discovery heuristics — pure unit tests, no network."""
from __future__ import annotations
from govcrawler.api.admin.discover_html import (
_column_id_from_path,
_has_article_links,
discover_html_sub_columns,
)
def test_column_id_from_path():
assert _column_id_from_path("/zwgk/zfwj/gfxwj/index.html") == "gfxwj"
assert _column_id_from_path("/zwgk/qxyw/") == "qxyw"
assert _column_id_from_path("/zwgk/") == "zwgk"
assert _column_id_from_path("/foo-bar/") == "foo_bar"
assert _column_id_from_path("/") is None
def test_has_article_links_threshold():
html = """
a
b
c
about
"""
assert _has_article_links(html) is True
assert _has_article_links(html, threshold=10) is False
assert _has_article_links("") is False
def test_discover_html_sub_columns_includes_index_html_children():
"""Anchors pointing at deeper /index.html under the same path qualify."""
html = """
政府文件
通知公告
政务要闻
"""
out = discover_html_sub_columns("https://x.example/zwgk/", html)
cids = sorted(c["column_id"] for c in out)
assert cids == ["qxyw", "tzgg", "zfwj"]
def test_discover_html_sub_columns_rejects_external_and_articles_and_self():
html = """
external
article
self
legit
duplicate
更多
non-index html leaf
"""
out = discover_html_sub_columns("https://x.example/zwgk/", html)
cids = [c["column_id"] for c in out]
# legit appears once, none of the rejects make it through
assert cids == ["legit"]
def test_discover_html_sub_columns_rejects_outside_subtree():
"""Links pointing above the portal path (e.g. site nav) are rejected."""
html = """
关于
inside
"""
out = discover_html_sub_columns("https://x.example/zwgk/", html)
cids = [c["column_id"] for c in out]
assert cids == ["inside"]